html-conduit 1.3.1 → 1.3.2

raw patch · 4 files changed

+30/−4 lines, 4 filesPVP ok

version bump matches the API change (PVP)

API changes (from Hackage documentation)

Files

ChangeLog.md +4/−0
html-conduit.cabal +1/−1
src/Text/HTML/TagStream.hs +1/−1
test/main.hs +24/−2

ChangeLog.md view

@@ -1,3 +1,7 @@+## 1.3.2++* Fix a bug that was removing `<` symbols in script tags.+ ## 1.3.1  * Inline tagstream-conduit for entity decoding in attribute value bug

html-conduit.cabal view

@@ -1,5 +1,5 @@ Name:                html-conduit-Version:             1.3.1+Version:             1.3.2 Synopsis:            Parse HTML documents using xml-conduit datatypes. Description:         This package uses tagstream-conduit for its parser. It automatically balances mismatched tags, so that there shouldn't be any parse failures. It does not handle a full HTML document rendering, such as adding missing html and head tags. Note that, since version 1.3.1, it uses an inlined copy of tagstream-conduit with entity decoding bugfixes applied. Homepage:            https://github.com/snoyberg/xml

src/Text/HTML/TagStream.hs view

@@ -167,7 +167,7 @@       chunk <- takeTill (== '<')       let acc' = acc <> B.fromText chunk           finish = pure [open, Text $ L.toStrict $ B.toLazyText acc', TagClose "script"]-          hasContent = (string "/script>" *> finish) <|> loop acc'+          hasContent = (string "/script>" *> finish) <|> loop (acc' <> "<")       (char '<' *> hasContent) <|> finish  tokens :: Parser [Token]

test/main.hs view

@@ -120,9 +120,9 @@      describe "script tags" $ do       it "ignores funny characters" $-        let html = "<script>hello > world</script>"+        let html = "<script>hello <> world</script>"             doc = X.Document (X.Prologue [] Nothing []) root []-            root = X.Element "script" Map.empty [X.NodeContent "hello > world"]+            root = X.Element "script" Map.empty [X.NodeContent "hello <> world"]          in H.parseLBS html @?= doc        {-@@ -152,3 +152,25 @@          in H.parseLBS html @?= doc      prop "parses all random input" $ \strs -> void $ evaluate $!! H.parseSTChunks $ map T.pack strs++    describe "#128 entities cut off" $ do+      it "reported issue" $ do+        let html = "<a href=\"https://example.com?a=b&#038;c=d\">link</a>"+            doc = X.Document (X.Prologue [] Nothing []) root []+            root = X.Element+              "a"+              (Map.singleton "href" "https://example.com?a=b&c=d")+              [X.NodeContent "link"]+         in H.parseLBS html @?= doc++      it "from test suite" $ do+        let html = "<a class=\"u-url\" href=\"https://secure.gravatar.com/avatar/947b5f3f323da0ef785b6f02d9c265d6?s=96&#038;d=blank&#038;r=g\">link</a>"+            doc = X.Document (X.Prologue [] Nothing []) root []+            root = X.Element+              "a"+              (Map.fromList+                  [ ("href", "https://secure.gravatar.com/avatar/947b5f3f323da0ef785b6f02d9c265d6?s=96&d=blank&r=g")+                  , ("class", "u-url")+                  ])+              [X.NodeContent "link"]+         in H.parseLBS html @?= doc